This study tests four hypotheses related to forest clearing over time in Ecuador's northern Amazon: (1) a larger increase in population over time on a farm (finca) leads to more deforestation; (2) rates of forest clearing surrounding four primary reference communities differ (spatial heterogeneity); (3) fincas farther from towns/communities experience lower rates of forest clearing over time; and (4) forest clearing differs by finca settlement cohort, viz., by year of establishment of the finca. In this paper, we examine the relationship between forest clearing and key variables over time, and compare three statistical models-OLS, random effects, and spatial regression-to test hypotheses. Descriptive analyses indicate that 7-15% of forest area was cleared on fincas between 1990 and 1999; that more recently established fincas experienced more rapid forest clearing; and that population size and forest clearing are both related to distance from a major community. Controlling for key variables, model results indicate that an increase in population size is significantly related to more forest clearing; rates of forest clearing around the four major communities are not significantly different; distances separating fincas and communities are not significantly related to deforestation; and deforestation rates are higher among more recently established fincas. Key policy implications include the importance of reducing population growth and momentum through measures such as improving information about and provision of family planning services; increasing the low level of girls education to delay and reduce fertility; and expanding credit and agricultural extension services to increase agricultural intensification.